Mintong Kang

About Me

I am Mintong Kang, a fourth-year Ph.D. student at UIUC CS advised by Prof. Bo Li. My research interest lies in trustworthy machine learning and AI safety. I am interested in uncovering the vulnerability of advanced ML models and developing certifiable defense mechanisms to safeguard their universal deployments. I am recently working on the trustworthiness of multi-modal models (VLM, audio/video LLMs) and LLM agent systems.

Before that, I got the bachelor of engineering degree from the CS department of Zhejiang University. I work with Prof. Xi Li at DCD Lab @ Zhejiang University. I also luckily work with Prof. Alan L. Yuille at CCVL Lab @ Johns Hopkins University.

News

• [Sep. 2025] Two first-authored papers accepted to NeurIPS 2025.
• [Jun. 2025] Started research internship at NVIDIA Research.
• [Jan. 2025] Four papers accepted to ICLR 2025.

Selected Publications

• C-SafeGen: Certified Safe LLM Generation with Claim-Based Streaming Guardrails
  Mintong Kang, Zhaorun Chen, Bo Li
  [NeurIPS 2025] (Thirty-Ninth Annual Conference on Neural Information Processing Systems)
  [PDF] [Code]

• GuardSet-X: Massive Multi-Domain Safety Policy-Grounded Guardrail Dataset
  Mintong Kang, Zhaorun Chen, Chejian Xu, Jiawei Zhang, Chengquan Guo, Minzhou Pan, Ivan Revilla, Yu Sun, Bo Li
  [NeurIPS 2025] (Thirty-Ninth Annual Conference on Neural Information Processing Systems)
  [PDF] [Code]

• AdvAgent: Controllable Blackbox Red-teaming on Web Agents
  Chejian Xu, Mintong Kang, Jiawei Zhang, Zeyi Liao, Lingbo Mo, Mengqi Yuan, Huan Sun, Bo Li
  [ICML 2025] (Forty-Second International Conference on Machine Learning)
  [PDF] [Code]

• ShieldAgent: Shielding Agents via Verifiable Safety Policy Reasoning
  Zhaorun Chen, Mintong Kang, Bo Li
  [ICML 2025] (Forty-Second International Conference on Machine Learning)
  [PDF] [Code]

• $R^2$-Guard: Robust Reasoning Enabled LLM Guardrail via Knowledge-Enhanced Logical Reasoning
  Mintong Kang, Bo Li
  [ICLR 2025] (Thirteenth International Conference on Learning Representations)
  [PDF] [Code]

• AdvWave: Stealthy Adversarial Jailbreak Attack against Large Audio-Language Models
  Mintong Kang, Chejian Xu, Bo Li
  [ICLR 2025] (Thirteenth International Conference on Learning Representations)
  [PDF] [Code]

• MMDT: Decoding the Trustworthiness and Safety of Multimodal Foundation Models
  Chejian Xu*, Jiawei Zhang*, Zhaorun Chen*, Chulin Xie*, Mintong Kang*, Zhuowen Yuan*, Zidi Xiong*, Chenhui Zhang, Lingzhi Yuan, Yi Zeng, Peiyang Xu, Chengquan Guo, Andy Zhou, Jeffrey Ziwei Tan, Zhun Wang, Alexander Xiong, Xuandong Zhao, Yu Gai, Francesco Pinto, Yujin Potter, Zhen Xiang, Zinan Lin, Dan Hendrycks, Dawn Song, Bo Li
  [ICLR 2025] (Thirteenth International Conference on Learning Representations)
  [PDF] [Code]

• EIA: Environmental Injection Attack on Generalist Web Agents for Privacy Leakage
  Zeyi Liao*, Lingbo Mo*, Chejian Xu, Mintong Kang, Jiawei Zhang, Chaowei Xiao, Yuan Tian, Bo Li, Huan Sun
  [ICLR 2025] (Thirteenth International Conference on Learning Representations)
  [PDF] [Code]

• C-RAG: Certified Generation Risks for Retrieval-Augmented Language Models
  Mintong Kang, Nezihe Merve Gürel, Ning Yu, Dawn Song, Bo Li
  [ICML 2024] (Forty-first International Conference on Machine Learning)
  [PDF] [Code]

• Certifiably Byzantine-Robust Federated Conformal Prediction
  Mintong Kang, Zhen Lin, Jimeng Sun, Cao Xiao, Bo Li
  [ICML 2024] (Forty-first International Conference on Machine Learning)
  [PDF] [Code]

• COLEP: Certifiably Robust Learning-Reasoning Conformal Prediction Via Probablistic Circuits
  Mintong Kang, Nezihe Merve Gürel, Linyi Li, Bo Li
  [ICLR 2024] (Twelfth International Conference on Learning Representations)
  [PDF] [Code]

• DecodingTrust: A Comprehensive Assessment of Trustworthiness in GPT Models
  Boxin Wang*, Weixin Chen*, Hengzhi Pei*, Chulin Xie*, Mintong Kang*, Chenhui Zhang*, Chejian Xu, Zidi Xiong, Ritik Dutta, Rylan Schaeffer, Sang T. Truong, Simran Arora, Mantas Mazeika, Dan Hendrycks, Zinan Lin, Yu Cheng, Sanmi Koyejo, Dawn Song, Bo Li
  [NeurIPS 2023] (Outstanding Paper Award, Oral) (Thirty-seventh Conference on Neural Information Processing Systems Datasets and Benchmarks Track)
  [PDF] [Code]

• DiffAttack: Evasion Attacks Against Diffusion-Based Adversarial Purification
  Mintong Kang, Dawn Song, Bo Li
  [NeurIPS 2023] (Thirty-seventh Conference on Neural Information Processing Systems)
  [PDF] [Code]

• Certifying Some Distributional Fairness with Subpopulation Decomposition
  Mintong Kang*, Linyi Li*, Maurice Weber, Yang Liu, Ce Zhang, Bo Li
  [NeurIPS 2022] (Spotlight) (Thirty-sixth Conference on Neural Information Processing Systems)
  [PDF] [Code]

Services

• Conference Reviewers: ICML 2022-2024; NeurIPS 2022-2024; ICLR 2024; CVPR 2023-2024; KDD 2023; AAAI 2025.
• Workshop organizer: KLR@ICML 2023.

Powered by Jekyll and Minimal Light theme.